|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Google Scholar output for 2q9k | |
|---|---|
| 1. Decision-making in structure solution using Bayesian estimates of map quality: the PHENIX AutoSol wizard | |
| TC Terwilliger, PD Adams, RJ Read - Section D: Biological , 2009 - scripts.iucr.org | |
| 2. Ligands in PSI structures | |
| A Kumar, HJ Chiu, HL Axelrod, A Morse - Section F: Structural , 2010 - scripts.iucr.org | |
| 3. A novel approach to studying the structural and functional properties of proteins with unknown functions | |
| MA Gorbacheva, AG Yarosh, PV Dorovatovskii - Russian Journal of , 2012 - Springer | |
| 4. The structure of a Xanthomonas general stress protein involved in citrus canker reveals its flavin-binding property | |
| E Hilario, Y Li, D Niks, L Fan - Acta Crystallographica Section D: , 2012 - scripts.iucr.org | |
The Exig_1997 gene from Exiguobacterium sibiricum 255-15 is a distant member of a pyridoxamine 5'-phosphate oxidase (PFAM:PF01243, cl00381) family, which mostly (but not exclusively) groups enzymes that catalyze the oxidation of pyridoxamine-5-P (PMP) and pyridoxine-5-P (PNP) to pyridoxal-5-phosphate (PLP).
The 2q9k structure adopts a beta roll (2q9k-CATH) fold. There is high structural similarity to several oxidoreductases like 1r1z [Ref] and PSI target PDB:2OU5. Interestingly, both HHpred and FFAS match it with other atypical members of the pyridoxamine 5'-phosphate oxidase family - HHpred with PDB:2e83 (chain A), PDB:3f7e (chain A) and FFAS to another JCSG structure, PDB:2htd. DALI top hits are with a predicted flavin nucleotide binding protein PDB:2htd (Z=12), the PNPO-related PDB:3f7e (Z=10), and the flavin mononucleotide (FMN) binding proteins PDB:1wlk and PDB:3in6 (Z=10).
The 2q9k structure contains electron density which could not be assigned a specific ligand and is annotated as an UNL ligand in the PDB deposition. Eleven atoms were identified - looking at distances and the electron density reveals that some of the atoms could be water molecules (O4 and O5) and O10 has very weak density, indicating that this atom does not exist. We emphasize that the position of this unknown ligand is in the binding pocket of FMN and PLP, as shown by structure comparison to 1g78, the X-ray structure of Escherichia coli pyridoxine 5'-phosphate oxidase complexed with PLP and FMN. However, the orientation and the size of this ligand cannot be matched with neither PLP nor FMN. An analysis of PLP binding residues in 1g78 [Ref] on a structure superposition with 2q9k does not reveal any conservation of binding. However, the structure analysis predicts that the FMN binding is still active, as shown in Figure 1.
Figure 1: Superposition of 2q9k (blue) with 1g78 (green). Structurally equivalent residues are colored red (2q9k) and orange (1g78). Superposition was made with TopMatch. The FMN cofactor is highlighted as a white stick model with oxygen in red, nitrogen in blue and phosphorus in pink. The PLP substrate is also shown as a stick model in beige. The atoms of the unknown ligand are shown as small grey spheres. Notice that these atoms are almost perpendicular to the planes of FMN and PLP. In 1g78 a critical residue for binding FMN is Arg-67, shown on the right side of the phosphate group in the top area of the figure. In 2q9k this is replaced by a His-38 residue, providing the same binding as an arginine. On the left side there is a conserved serine (Ser-59 in 2q9k and Ser-87 in 1g78) providing a hydroxyl group for binding the phosphate group. The beta strand to the right of the ligand also interacts via hydrogen bonds. The 2q9k structure is lacking interacting residues on the left side shown in green. It may therefore be the case that the ligand adopts a different orientation, with the phosphate group being fixed by Ser-59 and His-38.